SYSTEMS AND METHODS FOR SYNCHRONIZATION EVENT BUILDING AND/OR COLLAPSING BY A SYNCHRONIZATION COMPONENT OF A CLOUD-BASED PLATFORM

• US 20140372376A1
• Filed: 06/13/2014
• Published: 12/18/2014
• Est. Priority Date: 06/13/2013
• Status: Active Grant

• Current Assignee: Box Incorporated
• Original Assignee: Box Incorporated
• Inventors: Smith, Benjamin Campbell, Mackenzie, David, Parmar, Kunal, Potrebic, Peter, Lu, Yiming
• Granted Patent: US 9,633,037 B2
• Time in Patent Office: Days
• US Class Current: 707/624
• CPC Class Codes: G06F 16/178 Techniques for file synchronisation in file systems
• CPC Class Codes: G06F 16/184 implemented as replicated file system
• CPC Class Codes: G06F 16/273 Asynchronous replication or reconciliation

Abstract

Techniques for monitoring local and/or remote file systems by a synchronization component (e.g., client/server) of a cloud-based platform are disclosed. In some embodiments, a method of building synchronization events by a synchronization component (e.g., a synchronization server/client) includes obtaining a set of items that have been changed and their new states and retrieving last known states of the set of items that are stored in a reference snapshot inside a filesystem scanner. The method further includes generating differences between the new states and the last known states of the set of items as item changes and utilizing information provided by the item changes to translate the item changes into synchronization events for execution on the opposing file system. A method of handling failed synchronization events by a synchronization component of the cloud-based platform by collapsing a subsequent event with the failed synchronization event is also disclosed.

Claims

1. A method of building synchronization events by a synchronization component associated with a cloud-based platform, the method comprising:

obtaining a set of items that have been changed and new states of the set of items;

retrieving last known states of the set of items, wherein the last known states of the set of items are stored in a reference snapshot inside a filesystem scanner;

generating differences between the new states and the last known states of the set of items as item changes;

utilizing information provided by the item changes to translate the item changes into synchronization events.

2. The method of claim 1, further comprising receiving a notification of a change in a folder and in response scanning the folder to obtain the set of items that have been changed.

3. The method of claim 1, further comprising:

detecting event dependencies between the item changes;

ordering the item changes based on the detected event dependencies to allow the new state brought about by applying each item change is consistent with a filesystem.

4. The method of claim 1, further comprising:

checking each item change against a snapshot of a filesystem monitored by the synchronization component to determine if the new state brought about by the item change is consistent with the filesystem.

5. The method of claim 4, further comprising:

if the new state brought about by the item change is not consistent with the filesystem, buffering the item change until another item change alters the snapshot of the filesystem.

6. The method of claim 4, further comprising:

if the new state brought about by the item change is consistent with the filesystem, providing the item change to a filter pipeline for generating the synchronization events.

7. The method of claim 1, wherein the synchronization events include actions arranged in a predetermined order for execution on an opposing file system, wherein the predetermined order allows the synchronization events to be executed without generating any inconsistencies.

8. The method of claim 1, wherein the synchronization component is one of a synchronization client on a device or a synchronization server remote from the device, each associated with a respective filesystem.

9. A method of generating synchronization events associated with a synchronization component of a cloud-based platform, the method comprising:

detecting that a folder on a filesystem has been changed;

generating a new snapshot of the filesystem that is consistent;

retrieving a reference snapshot of the filesystem;

generating item changes by differencing the new snapshot from the reference snapshot; and

using the item changes to generate the synchronization events for execution on an opposing file system.

10. The method of claim 9, further comprising:

updating the reference snapshot using the new snapshot.

11. The method of claim 9, further comprising:

checking each item change against a current snapshot of the filesystem to determine an order for executing the item changes.

12. The method of claim 11, further comprising:

generating raw events in the determined order as executable actions to be applied to the opposing filesystem without generating inconsistencies.

13. The method of claim 12, further comprising:

processing the raw events to generate the synchronization events for execution.

14. The method of claim 13, wherein processing the raw events includes translating file system identifiers of items associated with the item changes to logical identifiers.

15. The method of claim 9, wherein differencing the new snapshot from the reference snapshot further comprises:

generating a list of all item identifiers that are in the reference snapshot;

processing each item in the new snapshot to obtain item changes for items that have different item states in the reference snapshot;

removing all items involved in the processing from the list; and

obtaining item changes with delete set to true for all items remaining in the list.

16. A system, comprising:

means for generating file system notifications, wherein the notifications include information about files or folders that have been changed;

means for determining item changes based on a consistent snapshot of a folder whose children have changed and a reference snapshot of the folder in response to the file system notifications;

means for generating raw events by ordering the item changes and producing a set of executable actions; and

means for generating synchronization events by mapping file system identifiers to logical identifiers for execution on an opposing file system.

17. A method of handling failed synchronization events by a synchronization component of a cloud-based platform, comprising:

detecting a failed synchronization event associated with a synchronization item;

identifying a subsequent synchronization event associated with the synchronization item to collapse with the failed synchronization event; and

re-executing the collapsed synchronization event on an opposing file system.

18. The method of claim 17, further comprising:

determining a failure reason associated with the failed synchronization event; and

identifying the subsequent synchronization event based on the failure reason.

19. The method of claim 18, wherein when the failure reason includes name conflict, the subsequent synchronization event includes a rename event.

20. The method of claim 18, wherein when the failure reason includes missing synchronization item on a source filesystem, the subsequent synchronization event includes a delete event.

21. The method of claim 18, wherein when the failure reason includes change in the synchronization item when deleting the synchronization item on the opposing file system, the subsequent synchronization event includes a create event.

22. The method of claim 17, wherein collapsing the subsequent synchronization event with the failed synchronization event results in a single operation or no operation, wherein the single operation includes a create operation, a rename operation, a move operation or a delete operation.

23. A system for handling failed synchronization events associated with a cloud-based platform, comprising:

a memory;

a processor in communication with the memory and configured to execute instructions stored in the memory to;

detect a failed synchronization event associated with a synchronization item;

identify a subsequent synchronization event associated with the synchronization item to combine with the failed synchronization event to generate a modified synchronization event; and

re-execute the modified synchronization event on an opposing file system.